Safety Assessment Of Power Transmission Corridors In Forestry Area Based On Multi-source Data

High-voltage power lines(PLs)are bases that economy expands,but distributed widely in the field.There are kinds of disasters(e.g.wildfire,flood,lighting,etc.)that affect the steady operation of the PLs.Transmission failures were often caused by wildfires,or vegetation growing around the PL corridor.As one of the disasters,wildfire is a vital factor disturbing the power transmission system and the surrounding vegetation in the forestry area.On the one hand,the distance from PLs to the vegetation are so close,i.e.less than 100 meters,that could have influence on the power transmission by the heated atmosphere and smog.When it is burning,there could be a transmission fault owning to smoke particles,or even a damage of the foundation and a deformation of the transmission pylon,resulting in serious consequences.On the other hand,it would lead to contact between PL and trees with the growing of the vegetation near the PLs.Once a tree bends,breaks or falls due to winds,ice or other meteorological disasters,the contacted branch or a whole tree could be ignited due to discharge,which will lead to a forest fire.In this dissertation,a high-voltage transmission corridor is defined as a strip with a certain width(about 2-3km)formed by the overhead PLs,pylons,terrain,vegetation and other objects.The theoretical principle of using multitype remote sensing data to predict forest fire risk was discussed,and methods to assess the fire risk in transmission corridor were carried out.Additionally,as time and resources are limited,and the routine clearance inspection defending vegetation encroachment is a heavy task along all those PLs with a wide range,it is difficult to achieve reasonable and efficient allocation to finish all these tasks completely.The risk assessment can fortunately help making a scientific plan for routine inspection according to the PLs’ risk level,and making policies for different risks,so as to ensure the priority of critical inspection in high-risk areas and the gradual reduction of priority in other areas.The risk assessment data of PL corridor on the mesoscale and small scale would be applied to prevent the wildfire risk in PL corridor.The purpose of this research is to serve the safe operation of power grids,to improve the pertinence and system efficiency of PL inspection,and to provide data support and suggestions for protection of power transmission against disasters.The risk assessment of wildfire and vegetation obstacle in power corridor was deeply explored in this dissertation.Firstly,the risk of wildfire was assessed,with high risk areas recognized and potential vegetation obstacles around each PL identified.Secondly,the early warning to power grid can be alarmed according to the risk level and its possible location,providing accurate information for the security management of high-voltage PLs and decision support for the disaster prevention and the safety of grid.The specific research includes the following aspects:(1)Facing the frequently occurred wildfires around the forestry PL corridor,the risk assessment methodology of wildfire in the long large-scale transmission area was studied,in order to accurately assess the risk of wildfires,avoid the blindness of security inspection,and perform intelligent and targeted PL inspection.Considering various factors relevant to the fire risk,including surface vegetation condition,topography,humanity,season,meteorology and their correlation with the occurrence of wildfire,a fire risk index of PL transmission corridor(PC-FRI)is designed to determine the areas and time periods getting high risk of wildfire in the whole corridor and to judge areas that need urgent inspection.This would provide accurate data for the further routine inspection,and lay a foundation for the early warning of the potential disaster points of vegetation obstacles.(2)Based on the analysis of fire risk in the high-voltage PL corridor,an effective method to automatically extraction high-voltage power transmission object was developed to improve the efficiency and effect of unmanned aerial vehicle(UAV)lidar inspection data processing,and to improve the automation level of the data processing.The management of point cloud data was optimized by combining the spatial hash structure,and the point clouds were stored in voxel blocks hierarchically.So that the transmission targets could be extracted by the local distribution and dimensional characteristics of grid data,which improved the accuracy and efficiency of this threedimensional(3D)target extraction.On this basis,for the high-voltage power lines,pylons and vegetation,the 3D model in the power corridor scene was reconstructed.The UAV lidar inspection system was used to collect the corridor inspection data in the high-level fire risk area in this section.(3)An efficient algorithm was developed to evaluate the potential risk status of vegetation obstacles using the unmanned aerial vehicle lidar inspection data.For that the troubles in the safe operation of the power transmission system in the high-voltage PL corridor were usually caused by the growing trees that almost touched the PLs,the distances between trees and PLs was calculated.Moreover,on the micro scale,given the growth rate of the trees,the distances and risk status of vegetation obstacles could be achieved.Then,the risk level was judged and warning points alerted by this method.Finally,the effectiveness of these methods was verified and the fire risk model was applied in two high-voltage transmission corridors of Shaoguan Power Supply Bureau of Guangdong Province.